SUMMARY OF WORK During the last project period, we developed new, more powerful Monte Carlo simulation algorithms in order to model local control in skeletal muscle. As planned, these methods have now been adapted and extended to cardiac excitation-contraction coupling. It was necessary to develop a new gating scheme for the L-type calcium channel (also known as dihydropyridine receptor, DHPR), since none of the published gating schemes for this channel were consistent with the experimental data on both calcium-dependent inactivation of L-type current and gating charge movement. Preliminary results of stochastic simulation of cardiac ECC show that local control can, in fact, explain the gradation and stability of SR calcium release, and can model quantitatively the rate of SR calcium release measured in our laboratory using the new Oregon green/EGTA technique. The simulations also demonstrate the existence of several types of competition between different DHPR's to release SR calcium, which can explain previously puzzling features of the published experimental data on cardiac EC coupling from several laboratories. However, a new paradox has arisen: the model only demonstrates adequate local and global stability if we assume the existence of a sufficiently powerful inactivation process of the cardiac ryanodine receptor, as well as cooperative activation of this channel by more than one calcium ion--features that have not been convincingly demonstrated in studies of isolated cardiac RyR gating in lipid bilayers. This has led to the hypothesis that the gating of RyR's in vivo is significantly different from that in vitro, possibly as a result of allosteric interactions between the foot processes of neighboring channels.